Arachidonic acid metabolism in metabolic dysfunction–associated steatotic liver disease and liver fibrosis

Abstract

Metabolic dysfunction–associated steatotic liver disease (MASLD) is the most prevalent chronic liver disorder globally, affecting 30% of the population and causing a significant healthcare burden due to its increasing incidence and limited therapeutic options. Arachidonic acid (AA) is a key bioactive lipid precursor that generates eicosanoids, such as prostaglandins, leukotrienes, and epoxyeicosatrienoic acids, via 3 distinct enzymatic pathways: cyclooxygenase, lipoxygenase, and cytochrome P450. Emerging evidence indicates that AA-derived metabolites and pathway factors contribute to the progression and severity of MASLD and liver fibrosis. This review systematically summarizes the pathophysiological roles of AA metabolism in MASLD and liver fibrosis, focusing on mechanisms involving lipid accumulation, liver inflammation, fibrogenesis, and related cellular processes. In addition, we discuss potential therapeutic targets within the AA metabolic pathway in MASLD and liver fibrosis, highlighting emerging clinical advances targeting AA metabolites and pathway factors to improve these pathological conditions.

INTRODUCTION

Metabolic dysfunction–associated steatotic liver disease (MASLD), formerly known as non-alcoholic fatty liver disease (NAFLD), is emerging as a leading cause of chronic liver disease, imposing a significant global health and economic burden.¹,²,³ MASLD currently affects ~30% of the global population, with its prevalence increasing to 590 million individuals by 2035.²,⁴ MASLD encompasses a clinical spectrum ranging from simple steatosis (metabolic dysfunction–associated steatotic liver, MASL) to metabolic dysfunction–associated steatohepatitis (MASH).⁵ While MASL is typically reversible with early intervention, progressive disease MASH manifests as worsening steatosis, significant inflammatory infiltration, and hepatocyte ballooning, ultimately leading to liver cirrhosis, HCC, and liver failure.⁶,⁷ Fibrotic progression, mediated by chronic inflammation and excessive extracellular matrix deposition in the Disse space, represents a common pathological endpoint across various chronic liver diseases, including MASLD.⁸ Liver fibrosis/cirrhosis often results in portal hypertension accompanied by aberrant angiogenesis.⁹,¹⁰ Notably, MASLD has also emerged as an independent risk factor for extrahepatic complications, particularly cardiovascular diseases and malignancies.¹¹ However, the substantial clinical heterogeneity and mechanistic complexity of MASLD continue to pose significant challenges for therapeutic development.¹²

Arachidonic acid (AA), an ω-6 polyunsaturated fatty acid (PUFA), is predominantly esterified in mammalian cell membrane phospholipids and plays a crucial role in maintaining membrane fluidity and cellular function.¹³,¹⁴ Upon cellular stimulation, AA is released via phospholipase-mediated hydrolysis and metabolized through 3 key enzymatic pathways: cyclooxygenases (COXs) produce prostaglandins (PGs) and thromboxanes (TXs); lipoxygenases (LOXs) generate leukotrienes (LTs), lipoxins (LXs), and hydroxyeicosatetraenoic acids (HETEs); and cytochrome P450 (CYP) enzymes catalyze the formation of epoxyeicosatrienoic acids (EETs) and additional HETE derivatives.¹⁴,¹⁵ These eicosanoids primarily exert their biological effects through G protein-coupled receptor (GPR) signaling (Figure 1).¹⁶ Thus, emerging evidence highlights the crucial role of AA and its metabolites in the pathogenesis of MASLD and liver fibrosis.¹⁶,¹⁷,¹⁸

FIGURE 1.

Overview of arachidonic acid metabolic pathways. Upon cellular stimulation, phospholipases catalyze the conversion of membrane phospholipids to AA. (1) The COX pathway, initiated by COX-1/2, leads to the production of PGs, PGI2, and TXs. (2) The LOX pathway generates HPETEs and HETEs, which further transform into various eicosanoids. (3) The CYP pathway catalyzes the conversion of AA into EETs and DHETs. These eicosanoids exert their biological effects by binding to GPRs. Created in BioRender. Ma, Y. (2025) https://BioRender.com/yruu09r. Abbreviations: AA, arachidonic acid; ALX, lipoxin A4 receptor; COX, cyclooxygenase; CysLT1R, cysteinyl leukotriene receptor 1; CYP, cytochrome P450; DHET, dihydroxyeicosatrienoic acid; DP, PGD2 receptor; EET, epoxyeicosatrienoic acid; EP, PGE2 receptor; EX, eoxin; FP, PGF2α receptor; GPR, G protein-coupled receptor; GPX, lutathione peroxidase; 5-HEDH, 5-hydroxyeicosanoid dehydrogenase; HETE, hydroxyeicosatetraenoic acid; HPETE, hydroperoxyeicosatetraenoic acid; HX, hepoxilin; IP, PGI2 receptor; LBT1/2, leukotriene B4 receptor 1/2; LT, leukotriene; LOX, lipoxygenase; LX, lipoxin; oxo-ETE, oxo-eicosatetraenoic acid; PG, prostaglandins; 15-PGDH, 15-hydroxyprostaglandin dehydrogenase; PGI2, prostacyclin; sEH, soluble epoxide hydrolase; TrX, trioxillin; TX, thromboxane; TP, TXA2 receptor.

Briefly, dysregulated AA metabolism disrupts glucose and lipid homeostasis, thereby exacerbating pro-inflammatory responses and fibrotic progression.¹⁶,¹⁷,¹⁸,¹⁹ AA-derived mediators orchestrate key pathophysiological mechanisms by modulation of apoptosis,²⁰ pyroptosis,²¹ autophagy,²²,²³,²⁴ cellular senescence,²⁴,²⁵ and ferroptosis.²⁶ In addition, the bidirectional interplay between AA metabolic dysregulation and gut microbiota dysbiosis amplifies disease progression via gut–liver axis interactions.²⁷ These findings position the AA metabolic network as a central regulator in initiating and progressing MASLD and liver fibrosis. Thus, this review systematically summarizes the current understanding of the AA-mediated mechanisms underlying hepatic lipid accumulation, inflammatory activation, and extracellular matrix remodeling in MASLD and liver fibrosis. We further discuss potential therapeutic targets within the AA metabolic pathway in MASLD and liver fibrosis, which may offer potential interventions for MASLD management and antifibrotic therapy.

THE AA METABOLIC PATHWAY

AA, typically referring to all-cis-5,8,11,14-eicosatetraenoic acid, is a PUFA comprising 20 carbon atoms and 4 cis double bonds (Figure 1).²⁸ AA is derived from both endogenous and exogenous sources. Endogenously, AA is released from membrane phospholipids. While exogenously, AA is obtained through dietary components such as meat products, eggs, and fish.²⁸ In addition, AA can be biosynthesized from linoleic acid, an essential ω-6 PUFA primarily sourced from vegetable oils and walnuts.²⁸

The conversion of membrane phospholipids to AA is catalyzed by various phospholipases, including calcium-dependent cytoplasmic phospholipase A2 (cPLA2), calcium-independent phospholipase A2 (iPLA2), and secretory phospholipase A2 (sPLA2).¹⁵,¹⁶ Among these, cPLA2 preferentially hydrolyzes AA at the sn-2 position of phospholipids and is considered the dominant enzyme in mediating the production of pro-inflammatory eicosanoids.¹³ iPLA2α contributes to cellular homeostasis via the synthesis of specialized pro-resolving mediators (SPMs) and reacylation of free AA.²⁸ sPLA2 synergistically enhances AA mobilization when acting in concert with cPLA2.¹³,¹⁶ Besides PLA2, phospholipase C and phospholipase D also generate AA via intermediate products such as diacylglycerol (Figure 1).²⁸

The AA metabolic pathway serves as a critical source of both pro-inflammatory and anti-inflammatory mediators, playing a complex role in pathophysiological processes. AA metabolism has long been recognized for its pro-inflammatory role in the pathogenesis of MASLD.¹⁷ Emerging evidence also highlights its anti-inflammatory potential by suppressing the NOD-like receptor family pyrin domain-containing 3 (NLRP3) inflammasome.²⁹ SPMs, including LXs, resolvins (Rvs), protectins (PDs), and maresins (MaRs), limit inflammatory cell migration and tissue infiltration (Table 1).³⁴ Consistently, SPMs also switch the macrophage phenotype from the pro-inflammatory M1 phenotype to the pro-resolving M2 phenotype, downregulate pro-inflammatory mediators (eg, PGs and LTs), and upregulate anti-inflammatory mediators such as interleukin-10 (IL-10).³⁴ The delicate balance between the pro-inflammatory and anti-inflammatory metabolites of AA is crucial for maintaining physiological homeostasis. Disruption of this equilibrium, particularly through excessive production of pro-inflammatory eicosanoids, can lead to pathological manifestations, including MASLD and liver fibrosis. AA is released via phospholipase-mediated hydrolysis and metabolized by COXs, LOXs, and CYP enzymes. The detailed process is described below (Figure 1).

TABLE 1.

Main specialized pro-resolving mediators and their biosynthetic sources

Series	Mediators	Sources	Key catalytic enzymes
LXs30	LXA4, LXB4	AA	15-LOX, 5-LOX, acetylated-COX-2
Rvs31	RvD1, RvD2, RvD3, RvD4, RvD5, RvD6	DHA	15-LOX, 5-LOX, acetylated-COX-2
	RvE1, RvE2, RvE3, RvE4	EPA	Acetylated-COX-2, 5-LOX, 12/15-LOX, CYP450
PDs32	PD1	DHA	15-LOX
MaRs33	MaR1, MaR2	DHA	12-LOX, sEH

Abbreviations: AA, arachidonic acid; COX, cyclooxygenase; DHA, docosahexaenoic acid; EPA, eicosapentaenoic acid; LOX, lipoxygenase; LX, lipoxin; MaR, maresin; PD, protectin; Rv, resolvin; sEH, soluble epoxide hydrolase.

The COX signaling pathway

COXs, also known as prostaglandin G/H synthases, are the first enzymes involved in converting AA to prostaglandin H2 (PGH2; Figure 1). COXs also utilize alternative substrates, including eicosapentaenoic acid and dihomo-γ-linolenic acid, to produce PGH3 and PGH1 derivatives, respectively.³⁵ Despite structural differences within the active sites of the two COX isoforms, both COX-1 and COX-2 mediate the sequential conversion of AA to prostaglandin G2 (PGG2) and then to PGH2.³⁶ Subsequently, PGH2 is converted into a subclass of eicosanoids called prostanoids, including prostaglandin D2 (PGD2), prostaglandin E2 (PGE2), prostaglandin F2α (PGF2α), prostacyclin (PGI2), and thromboxane A2 (TXA2), through the actions of prostaglandin synthase and thromboxane synthase (Figure 1).³⁷ These lipid mediators, characterized by their prostanoic acid backbone, are rapidly synthesized and released in response to diverse stimuli to regulate both physiological homeostasis and pathological processes.³⁷,³⁸

COX-1

COX-1 (PTGS1), constitutively expressed in most cells, is the primary source for prostanoids involved in homeostatic functions.¹⁵ COX-1 catalyzes the conversion of AA to TXA2, PGI2, and other prostanoids, which mediate crucial signaling pathways after interacting with their corresponding receptors, the TXA2 receptor (TP) and the PGI2 receptor (IP; Figure 1).¹⁶,³⁹ Contrary to the traditional view that COX-1 expression is solely developmentally regulated, recent studies have shown that IL-4 can upregulate COX-1 expression in bone marrow-derived macrophages via post-transcriptional mechanisms involving the Fes–Akt–mTORC axis. Inhibition of this pathway impairs both COX-1 expression and M2 macrophage polarization.⁴⁰

COX-2

COX-2 (PTGS2), induced by inflammatory stimuli, hormones, and growth factors, is considered the primary source of prostanoid production in inflammation and proliferative diseases.¹⁵ Notably, COX-2 is also constitutively expressed in select regions including the brain, lung, thymus, gut and kidney.³⁶ COX-2 preferentially generates PGE2, PGD2, and PGF2α, which modulate inflammatory and metabolic responses through their cognate receptors (EP1-4, DP1-2, and FP; Figure 1).¹⁶,³⁹ Genetic deletion studies have revealed compensatory eicosanoid production between the COX-1 and COX-2 signaling pathways.³⁹

The LOX signaling pathway

LOXs are a family of non-heme iron-containing enzymes that are ubiquitously present across different species.⁴¹,⁴² In mammals, LOXs exhibit predominant activity toward AA and linoleic acid, with some isoforms also acting on α-linolenic acid, docosahexaenoic acid, and eicosapentaenoic acid.⁴³ These oxygenated PUFAs initiate signaling pathways by interacting with specific cell-surface receptors or are further metabolized into bioactive lipid mediators, including LTs, LXs, hepoxilins (HXs), eoxins (EXs), trioxillins (TrXs), Rvs, PDs, and MaRs. These mediators function as pro-inflammatory and anti-inflammatory mediators, as well as SPMs, participating in membrane restructuring, lipoprotein interactions, and the regulation of cellular redox states.⁴⁴

Humans possess 6 functional LOX genes encoding 6 arachidonate LOX (ALOX) enzymes: ALOX5, ALOX12, ALOX12B, ALOX15, ALOX15B, and ALOXE3. In mice, the corresponding functional LOX genes are Alox12, Alox12a, Alox15, Alox15b, Aloxe3, and Alox12e.⁴⁴ LOXs insert oxygen into AA, forming 5-, 12-, and 15-hydroperoxyeicosatetraenoic acid (HPETE), which are catalyzed by 5-LOX, 12-LOX, and 15-LOX, respectively. HPETEs can be reduced to HETEs by glutathione peroxidase or converted into biologically active compounds such as LTs, LXs, HXs, TrXs, and EXs (Figure 1).¹⁵,¹⁶

5-LOX

5-LOX catalyzes the formation of 5-HPETE from AA, which can be further metabolized to produce LTA4 and 5-HETE.²⁸ The unstable epoxide LTA4 is subsequently converted either to LTB4 by LTA4 hydrolase or to cysteinyl leukotrienes (cysLTs; LTC4, LTD4, and LTE4) by LTC4 synthase.¹⁵,¹⁶ LTA4 oxidation by 12-LOX and 15-LOX produces LXs, such as LXA4 and LXB4.³⁷ In addition, 5-HPETE can be reduced to 5-HETE, which can be further oxidized to 5-oxo-eicosatetraenoic acid by microsomal dehydrogenase in polymorphonuclear leukocytes (PMNs; Figure 1).⁴⁵ The catalytic activity of 5-LOX requires the 5-LOX-activating protein (FLAP), a membrane-spanning protein with three transmembrane domains that include LTC4 synthase and microsomal PGE2 synthase.¹⁵ 5-LOX plays a crucial role in immune regulation by generating both pro-inflammatory mediators (eg, HETEs, LTB4, and cysLTs) and anti-inflammatory mediators such as LXs.⁴⁴

LTB4 and cysLTs are potent inflammatory mediators that induce pain, fever, and inflammation when produced in excess.⁴⁶ LTB4 is a powerful chemotactic agent for human PMNs, inducing chemokinesis and chemotaxis.³⁸,⁴⁶ LTB4 exerts its effects by binding to the cell-surface receptors LTB4R1 (BLT1) and LTB4R2 (BLT2), triggering distinct intracellular signaling cascades.⁴⁷ By binding to the BLT1 and BLT2, LTB4 activates neutrophils and promotes neutrophil extracellular trap (NET) formation, which involves a specific form of cell death known as NETosis.⁴⁸

12-LOX/15-LOX

12-LOX, also known as platelet-type 12-LOX, catalyzes the formation of 12-HPETE from AA. Three 12-LOX isoforms have been identified based on their cellular expression: leukocytes, platelets, and epidermis.⁴⁴ Similarly, 15-LOX, also referred to as 15-LOX-1, oxygenates AA to produce 15-HPETE. 15-LOX is primarily expressed in blood cells (reticulocytes, macrophages, eosinophils, dendritic cells, and neutrophils) and in epithelial, endothelial, and fibroblast cells.⁴⁴ In rodents, leukocyte-type 12-LOX is considered an ortholog of ALOX15 and is designated Alox15. The human ALOX15, as well as rodent leukocyte-type 12-LOX and ALOX15, metabolize AA to both 12-HPETE and 15-HPETE. Both these enzymes share 73% amino acid similarity, a similar expression pattern, and largely overlap in their known biological effects, thus commonly referred to as 12-/15-LOX.⁴⁹ Both 12-HPETE and 15-HPETE are further metabolized into 12-HETE and 15-HETE, respectively, and into additional bioactive products such as LXs, HXs, TrXs, EXs, and 15-oxo-ETEs.¹⁵,¹⁶,⁴⁴ Specifically, 12-HPETE undergoes isomerization to HXs through rearrangement of the −OOH group.³⁷ The epoxide group of HXs is labile and undergoes hydrolysis under the action of epoxide hydrolases, forming TrXA3 or TrXB3, respectively.⁵⁰ In addition, the action of 15-LOX leads to the metabolism of 15-HPETE to EXA4, which is conjugated with glutathione, forming EXC4–EXE4 (Figure 1).³⁷ In addition, under the influence of 5-LOX on 15-HPETE, lipoxins LXA4 and LXB4 are also formed.³⁷ 12-HETE exerts its effects by binding to its specific receptor GPR31.⁵¹ On the other hand, 15-LOX-2 generates 15-HPETE, which is further metabolized into 8-HETE and 15-oxo-ETE.¹⁵

SPMs

SPMs, including LXs,³⁰ RVs,³¹ PDs,³² and MaRs,³³ are distinct lipid mediators that play crucial roles in promoting the resolution of inflammation, tissue regeneration, and restoring homeostasis (Table 1). AA-derived LXs, including LXA4 and LXB4, are the first class of SPMs.³⁴ SPMs are generated from PUFAs through transcellular reactions involving interactions between M2 macrophages or PMNs and tissues at the site of inflammation, in response to tissue damage and infection.⁵² SPMs exert their effects by limiting PMN migration and tissue infiltration, enhancing macrophage phagocytosis of microbes, and promoting efferocytosis of apoptotic cells and cellular debris. These actions collectively facilitate the switch of macrophage phenotypes from the pro-inflammatory M1 phenotype to the pro-resolving M2 phenotype.³⁴ However, recent studies have highlighted fundamental flaws in the commonly applied liquid chromatography–tandem mass spectrometry (LC–MS/MS) approach to SPM detection.⁵³,⁵⁴ Consequently, the evidence supporting the presence of SPMs in biological matrices and their role in inflammation resolution warrants careful reevaluation. The physiological relevance of SPMs as endogenous compounds produced in sufficient quantities to exert anti-inflammatory effects in vivo remains to be fully elucidated.

The CYP signaling pathway

CYP enzymes, primarily located in the liver, are a group of self-oxidizing heme proteins named for their specific absorption peak at 450 nm.¹⁶ CYP enzymes predominantly catalyze monooxygenase reactions, which are crucial for hormonal regulation, the metabolism of xenobiotics (eg, drugs, carcinogens, and environmental pollutants), and the biosynthesis or inactivation of endogenous compounds such as steroids, fat-soluble vitamins, fatty acids, and biogenic amines.¹⁵,⁵⁵ The expression and activity of CYP enzymes are regulated by hormones, growth factors, and transcription factors.¹⁵ The CYP enzyme family comprises numerous subclasses, among which epoxygenase and ω-hydroxylase activities are particularly important for the metabolism of AA. However, many CYP enzymes possess both epoxygenase and ω-hydroxylase functions, thereby generating a diverse spectrum of products.¹⁵ The functional versatility underscores the critical role of CYP enzymes in metabolizing AA and other substrates.

Epoxygenase

Epoxygenases, such as the CYP2J and CYP2C families, catalyze the formation of AA epoxides, namely EETs (5,6-EET; 8,9-EET; 11,12-EET; and 14,15-EET), which exhibit potent anti-inflammatory and hepatoprotective effects.³⁸,⁵⁶ In the liver, bioactive EETs are predominantly metabolized by soluble epoxide hydrolase (sEH) into the corresponding diols, or dihydroxyeicosatrienoic acids, which are less active than EETs (Figure 1).³⁸ The human liver generates high levels of 14,15-EET and 11,12-EET and expresses high levels of sEH.³⁸ EETs function as autocrine and paracrine signaling lipid mediators.⁵⁷ Extensive previous research has highlighted the metabolic benefits and cardioprotective effects of EETs, which are attributed to their biological effects on maintaining metabolic homeostasis and detoxification, alleviating vascular inflammation, and regulating vasoconstriction and dilation.¹⁶,⁵⁷

ω-Hydroxylase

The human fatty acid ω-hydroxylase gene family, CYP4 enzymes, metabolize AA into essential signaling molecules, including 20-HETE, EETs, and 12-hydroxyeicosatrienoic acid, and play essential roles in the homeostasis of fatty acids and fatty acid-derived biomolecules such as LTs and prostanoids.⁵⁸ The ω-hydroxylation of AA results in high 20-HETE levels in various tissues, accounting for 50%–75% of liver CYP-dependent AA metabolism.⁵⁹ Key enzymes converting AA to 20-HETE include CYP4A11, CYP4F2, and CYP4F3.⁵⁸ In addition to nonspecific receptors such as peroxisome proliferator-activated receptors (PPARs), recent evidence has identified GPR75 as a receptor for 20-HETE (Figure 1).⁵⁹,⁶⁰

ROLE OF AA METABOLISM IN MASLD AND LIVER FIBROSIS

MASLD is primarily driven by overnutrition, which leads to adipose depot expansion and ectopic fat accumulation. Subsequently, macrophage infiltration into visceral adipose tissue induces a pro-inflammatory state, resulting in insulin resistance. This insulin resistance impairs lipolysis, continuously delivering free fatty acids to the liver. Increased de novo lipogenesis overwhelms hepatic metabolic capacity.²,⁵ An imbalance in lipid metabolism induces the formation of lipotoxic lipids, which contribute to cellular stress, activate inflammasomes, and trigger cell apoptosis. These events subsequently trigger inflammation, tissue regeneration, and fibrogenesis.⁵,⁴⁴

AA and its metabolic derivatives play pivotal roles in the progression of MASLD and liver fibrosis.¹⁴,¹⁵ Experimental evidence has demonstrated that elevated AA levels exacerbate systemic inflammation via the toll-like receptor 4 (TLR4)–nuclear factor kappa-B (NF-κB) pathway, which disrupts hepatic lipid metabolism and glucose homeostasis, and induces insulin resistance.¹⁷,⁶¹,⁶²,⁶³,⁶⁴ Clinically, elevated serum AA concentrations have been observed in lean-phenotype MASLD patients compared with healthy controls.⁶² A cohort study in Hispanic youth with obesity suggests that dietary restriction of AA and PUFAs may be a key intervention for preventing MASLD progression, particularly in PNPLA3 risk allele carriers.⁶⁵ Moreover, elevated plasma levels of AA-derived oxylipins, such as PGF2α and LXB4, are characteristically observed in severe MASLD patients, suggesting concurrent activation of pro-inflammatory and resolution pathways.⁶⁶ Circulating PGF2α levels significantly distinguish MASLD from non-MASLD cases.⁶⁶ Overall, AA metabolism critically drives MASLD and liver fibrosis progression by disrupting lipid homeostasis, exacerbating inflammation, promoting fibrogenesis, and interacting with environmental factors. In the following section, we summarize the specific roles of AA metabolic enzymes and related metabolites in the initiation and progression of MASLD and liver fibrosis (Table 2).

TABLE 2.

Arachidonic acid metabolites involved in metabolic dysfunction–associated steatotic liver disease and liver fibrosis

Metabolites	Receptors	Functions	Reference
COX family
TXA2	TP	(i) Stimulates insulin resistance and lipogenesis; (ii) Promotes KC inflammation.	67
PGI2	IP	Inhibits MASH progression by modulating KC-mediated inflammation and reducing oxidative stress.	68
PGD2	DP1-2	Reduces lipogenic gene expression, alleviating MASL to MASH progression.	69
15d-PGJ2	PPARγ	(i) Induces insulin resistance; (ii) Inhibits oxidative stress.	70,71
PGE2	EP1-4	(i) Inhibits inflammation and hepatocyte apoptosis; (ii) Activates HSCs; (iii) Impairs gut barrier integrity; (iv) Promotes angiogenesis and portal hypertension.	20,72,73,74,75
PGF2α	FP	Contributes to portal hypertension.	38
LOX family
5-HETE	—	Exacerbates apoptosis.	76
12-HETE	GPR31	(i) Induces epithelial–mesenchymal transition and inflammatory response; (ii) Promotes HCC recurrence in fatty livers.	77
15-HETE	—	Promotes fatty acid uptake and lipid accumulation.	78
LTB4	BLT1, BLT2	(i) Promotes hepatocyte lipogenesis; (ii) Activates HSCs.	79,80
LTC4	CysLT1R, CysLT2R	Activates HSCs.	79
LXA4	ALX	Reduces inflammatory cytokines and increases anti-inflammatory IL-10.	81
CYP family
EET	—	(i) Attenuates inflammation and oxidative stress; (ii) Reduces fatty acid accumulation.	82,83,84
20-HETE	GPR75	Induces obesity-related insulin resistance and hyperlipidemia.	85

Abbreviations: 15d-PGJ2, 15-deoxy-Δ^12,14-prostaglandin J2; ALX, lipoxin A4 receptor; COX, cyclooxygenase; CYP, cytochrome P450; CysLT1R, cysteinyl leukotriene receptor 1; DP, PGD2 receptor; EET, epoxyeicosatrienoic acid; EP, PGE2 receptor; FP, PGF2α receptor; GPR, G protein-coupled receptor; HETE, hydroxyeicosatetraenoic acid; IP, PGI2 receptor; LBT1, leukotriene B4 receptor 1; LOX, lipoxygenase; LT, leukotriene; LX, lipoxin; PG, prostaglandin; PGI2, prostacyclin; PPARγ, peroxisome proliferator-activated receptor γ; TP, TXA2 receptor; TXA2, thromboxane A2.

Role of COXs and their products in MASLD and liver fibrosis

The liver comprises diverse cell populations, including parenchymal hepatocytes and non-parenchymal cells such as liver sinusoidal endothelial cells (LSECs), Kupffer cells (KCs; resident hepatic macrophages), and HSCs.⁶ Among these cell types, activated KCs and LSECs are the primary producers of prostanoids, generating substantial amounts of PGD2, PGE2, and TXB2.⁸⁶,⁸⁷ Notably, although COX-2 expression has been observed in fetal hepatocytes, hepatoma cell lines, and hepatocytes during liver regeneration,⁸⁸,⁸⁹ mature hepatocytes under normal physiological conditions lack detectable COX-2 activity.⁹⁰

COX-1

COX-1 is a housekeeping gene that is constitutively expressed in most cells. The COX-1 metabolites TXA2 and PGI2 are critical in regulating hepatic inflammation and lipid metabolism during MASLD progression.⁶⁷,⁶⁸,⁹¹,⁹² TXA2, which is elevated in individuals with diet-induced obesity, facilitates hepatic insulin resistance and lipogenesis by activating the Ca²⁺/calcium calmodulin-dependent kinase IIγ (CaMKIIγ)–protein kinase RNA-like endoplasmic reticulum kinase (PERK)–C/EBP homologous protein (CHOP)–tribbles-like protein 3 (TRB3) axis in hepatocytes and contributes to KC inflammation (Figure 2).⁶⁷ PGI2-IP signaling pathway protects against MASLD progression by modulating KC-mediated inflammatory responses and reducing oxidative stress, as evidenced by the severe steatohepatitis observed in IP receptor knockout mice fed a methionine- and choline-deficient (MCD) diet.⁶⁸ Although pharmacological inhibition of COX-1 (eg, genistein) attenuates diet-induced MASLD by reducing TXA2 biosynthesis,⁹¹ COX-1 inhibition in treating MASLD should be cautious because COX-1 is a housekeeping gene.

FIGURE 2.

The involvement of cyclooxygenases and their metabolites in metabolic dysfunction–associated steatotic liver disease and liver fibrosis. COX-1/2 and their metabolites are implicated in the pathogenesis of MASLD and liver fibrosis, exerting both detrimental and protective effects. The roles of different COX metabolites in MASLD and liver fibrosis progression are distinct, and the same COX metabolite may exert divergent effects depending on its specific binding to different receptors. Created in BioRender. Ma, Y. (2025) https://BioRender.com/m65xk0g. Abbreviations: COX, cyclooxygenase; DAMP, damage-associated molecular pattern; ER, endoplasmic reticulum; HSC, hepatic stellate cell; IL-1β, interleukin-1β; KEAP1, Kelch-like ECH-associated protein 1; NET, neutrophil extracellular trap; NLRP3, NOD-like receptor family pyrin domain-containing 3; NRF2, nuclear factor erythroid 2-related factor 2; PG, prostaglandin; PGI2, prostacyclin; 15d-PGJ2, 15-deoxy-Δ^12,14-prostaglandin J2; SASP, senescence-associated secretory phenotype; TX, thromboxane; WIPI2, WD repeat domain and phosphoinositide interacting.

COX-2

COX-2 is an inducible gene typically induced by inflammatory stimuli. In MASLD and liver fibrosis, the hepatic COX-2 expression and levels of PGs (eg, PGE2, PGD2) are significantly elevated.⁶⁹,⁷⁰,⁷²,⁹³ Thus, COX-2 is a good therapeutic target for MASLD and liver fibrosis. The selective COX-2 inhibitor celecoxib suppresses epithelial–mesenchymal transition by blocking the transforming growth factor (TGF)-β1/Smad signaling pathway and reducing inflammation and counteracts fibrogenesis through downregulating TGF-β, enhancing matrix metalloprotease-2 activity, and reducing collagen deposition.⁹³,⁹⁴ However, their precise roles in MALSD progression remain controversial.⁸⁸ Hepatic COX-2 expression protects against mitochondrial dysfunction, steatosis, obesity, and insulin resistance.⁸⁸,⁸⁹

PGE2, synthesized by PGE synthase, serves as a critical mediator of inflammation.¹⁵ In MASLD and liver fibrosis, the function of PGE2 depends on its specific binding to different receptors across diverse cell types under various extracellular stresses. In adipose tissue macrophages, the COX-2–PGE2–EP4 axis limits adipose tissue dysfunction and inflammation, thereby reducing fibrosis and abnormal angiogenesis.⁹⁵ In both human THP-1 and mouse bone marrow-derived macrophage polarization models, PGE2 promotes macrophage polarization by enhancing oxidative phosphorylation.⁹⁶ Meanwhile, neutrophils form NETs by extruding chromatin fibers, which regulate inflammatory responses. Notably, NETs stimulate TLR3, leading to COX-2 activation and PGE2 production, subsequently activating HSCs and promoting liver fibrosis (Figure 2).⁷² Celecoxib restores gut barrier integrity via COX-2–PGE2–EP2–extracellular signal-regulated kinase (ERK)-dependent upregulation of zonula occludens-1 (ZO-1) and E-cadherin, thereby mitigating gut-derived inflammation and alleviating liver fibrosis progression.⁷³,⁷⁴ Suppression of PGE2-related inflammation and subsequent lipid accumulation reduces the severity of MASLD and inhibits progression toward MASLD-associated HCC.⁹⁷ Collectively, as PGE2 is a key downstream molecule of the COX-2 signaling pathway, these studies highlight the multifaceted roles of the COX-2–PGE2–EP axis in MASLD and liver fibrosis and suggest it as a potential therapeutic target.

PGD2 is formed by enzymatic or non-enzymatic isomerization of PGH2 and plays a crucial role in lipid metabolism and inflammation.³⁸ Lipocalin-type PGDS (L-PGDS), a variant of PGD synthase (PGDS), produces PGD2 in metabolic tissue.¹⁶ The absence or inhibition of L-PGDS results in dyslipidemia, altered expression of lipogenesis genes, and accelerated progression of MASLD. The exacerbated lipid accumulation in hepatocytes due to L-PGDS deficiency can be mimicked by a DP1 receptor antagonist, suggesting that PGD2 exerts a protective effect against MASLD.⁶⁹

The COX-2-mediated production of 15-deoxy-Δ^12,14-prostaglandin J2 (15d‑PGJ2), a non-enzymatic degradation product of PGD2, may be implicated in the pathogenesis of MASLD and hepatic insulin resistance via PPARγ activation.⁷⁰ Notably, 15d-PGJ2 exerts its biological effects by covalently modifying cysteine residues in Kelch-like ECH-associated protein 1 (KEAP1), thereby triggering the release of nuclear factor erythroid 2-related factor 2 (NRF2), a master regulator that orchestrates antioxidant responses and cellular defense against oxidative stress (Figure 2).⁷¹ Given the critical role of the KEAP1–NRF2 pathway in regulating oxidative stress, further research is needed to elucidate its regulatory mechanisms when COX-2 is overexpressed, which may provide novel insights into the pathogenesis of MASLD.

COX-related apoptosis and pyroptosis

Apoptosis, a form of programmed cell death mediated by caspase-3/7 and receptor-interacting protein kinase-1/3 (RIPK-1/3), is a critical process in MASLD and liver fibrosis.²⁰,⁹⁸,⁹⁹,¹⁰⁰ During MASLD progression, caspase-dependent apoptosis is the predominant cell death in hepatocytes.¹⁰¹ In human MASH livers, the expression of key enzymes involved in PGE2 synthesis, COX-2, and microsomal PGE synthase-1, is significantly increased compared with controls and correlates with the MASLD activity score.²⁰ Consistently, celecoxib attenuates hepatocyte apoptosis by suppressing CHOP-mediated endoplasmic reticulum (ER) stress.¹⁰² Conversely, attenuation of PGE2 production enhances the tumor necrosis factor (TNF)-α–triggered inflammatory response and hepatocyte apoptosis in diet-induced MASLD (Figure 2).²⁰

Pyroptosis, a distinct type of programmed cell death characterized by nuclear pyknosis, DNA breakage, cell membrane rupture, membranous vesicle formation, and the release of reactive oxygen species (ROS) and inflammatory cytokines, results in hepatocyte dysfunction and contributes to liver inflammation and fibrosis.²¹,⁹⁹,¹⁰³ The upregulation of COX-2 in hepatocytes induces NLRP3 inflammasome-dependent pyroptosis. IL-1β and damage-associated molecular patterns released from pyroptotic hepatocytes stimulate STAT3-mediated pro-inflammatory signaling in KCs, forming regional immune response cascades that amplify NLRP3-dependent pyroptosis and liver inflammation (Figure 2).²¹

COX-related autophagy

Macroautophagy/autophagy, an evolutionarily conserved process for degrading macromolecules and damaged organelles within the cytosol, is crucial for maintaining hepatic and systemic homeostasis. Autophagy also functions as an anti-inflammatory pathway in MASLD and liver fibrosis.²²,²³,¹⁰⁴,¹⁰⁵ Hepatocytes exhibit increased basal autophagy levels due to their abundant lysosomes, which facilitate continuous turnover of cytosolic components. Impaired autophagy is implicated in the pathogenesis of MASLD and liver fibrosis.²²,²³,²⁴ Elevated hepatic COX-1 expression in patients with MASLD and diet-induced MASLD mouse models is essential for basal hepatocyte autophagy. Concurrently, MASH induced by liver-specific COX-1 deletion partially depends on WD repeat domain and phosphoinositide-interacting 2 (WIPI2)-mediated autophagy (Figure 2).²³ PTUPB, a dual sEH/COX-2 inhibitor, enhances hepatocyte autophagy by inhibiting the PI3K/AKT/mTOR pathway via SIRT1 activation, alleviating MASLD progression.²⁴ In addition, constitutive hepatocyte COX-2 expression increases autophagy, protecting against MASLD.⁸⁹

COX-related cellular senescence

Cellular senescence, characterized by irreversible proliferative arrest, is a critical driver of senescence-associated secretory phenotypes (SASPs).¹⁰⁶ COX-2 is a key mediator of cellular senescence.¹⁸ During hepatocyte senescence, COX-2 critically regulates the SASP via an autocrine feedback mechanism involving the PGE2–EP4 signaling pathway, which modulates immune microenvironment dynamics and facilitates senescence surveillance (Figure 2).²⁵ PTUPB suppresses cellular senescence and MASLD pathology by enhancing hepatocyte autophagy.²⁴ Thus, COX-2-mediated cellular senescence and the associated SASP significantly contribute to the pathogenesis of MASLD, suggesting the potential therapeutic benefits of targeting the COX-2 signaling pathway.

COX-related angiogenesis and portal hypertension

COX-related angiogenesis plays a crucial role in the progression of liver fibrosis and portal hypertension.⁷⁵,¹⁰⁷,¹⁰⁸ LSECs, the most abundant non-parenchymal liver cells, are characterized by transcellular pores and a lack of basement membrane.¹⁰⁸ Endothelial COX-2 promotes intrahepatic angiogenesis and fibrogenesis via hypoxia-inducible factor-1α (HIF-1α)–vascular endothelial growth factor (VEGF) signaling pathway in a cirrhotic rat model (Figure 2).⁷⁵ Celecoxib mitigates oxidative stress and angiogenesis by targeting NADPH oxidase 1/4 (NOX1/4)-derived ROS in LSECs via the COX-2–PGE2–EP2–ERK axis.¹⁰⁷

Disorders of LSEC function and LSEC capillarization significantly increase intrahepatic vascular resistance, contributing to portal hypertension.¹⁰⁸ The vasoconstrictive activities of TXA2, PGD2, and PGF2α increase portal pressure, whereas PGE2 and PGH2 produce the opposite effect.³⁸ Increased TXA2 production, mainly due to upregulated COX-1 in cirrhotic liver LSECs, contributes to hepatic endothelial dysfunction and portal hypertension.⁹² Celecoxib normalizes TXA2-induced increased transhepatic pressure gradients in MASLD.⁷⁵,¹⁰⁹

The protective role of constitutive hepatocyte-specific expression of human COX-2 in MASLD and liver fibrosis

Althoug COX-2 has harmful effects, it may also protect against MASLD progression. Constitutive hepatocyte-specific expression of human COX-2 (hCOX-2) protects against high-fat diet (HFD)-induced adiposity, inflammation, and insulin resistance, thereby mitigating the progression of MASLD.⁸⁸,⁸⁹ Moreover, hCOX-2 attenuates MASH and fibrosis by reducing inflammation, oxidative stress, and apoptosis, and by modulating HSC activation.¹¹⁰ Mechanistically, hepatocyte COX-2 expression restricts HSC activation by downregulating microRNA-23a-5p and microRNA-28a-5p, thereby reducing the expression of α-SMA and promoting HSC apoptosis.¹¹¹ These findings suggest that constitutive hepatocyte-specific hCOX-2 expression may confer hepatoprotective effects in MASLD and liver fibrosis and warrant caution with COX inhibitor use in affected patients.

In summary, COX-2 may be a double-edged sword in the development of MASLD and liver fibrosis. These controversial results of the COX-2–PGE2 axis in MASLD and liver fibrosis may partially be attributed to the different animal models and the amount of PGE2 and COX-2 inhibitors used. Presumably, the effect of COX-2 on MASLD and liver fibrosis may vary in different liver cells. Further investigations may also need to clarify the specific source of COX-2 in liver cells.

Role of LOXs and their products in MASLD and liver fibrosis

LOX enzymes can catalyze AA, including 5-LOX, 12-LOX (platelet-type 12-LOX), and 12-/15-LOX (leukocyte-type 12-LOX). These LOX enzymes, especially LOX-related ferroptosis, also contribute to the pathogenesis of MASLD and liver fibrosis.

5-LOX

5-LOX expression is elevated in MASLD and liver fibrosis.⁷⁶,⁷⁹,¹¹² In hepatocytes, increased 5-LOX expression and the subsequent production of LTB4 and 5-HETE impair cell viability and exacerbate TNF-α-induced apoptosis by inhibiting NF-κB activity (Figure 3).⁷⁶ Compared with hyperlipidemia-prone apolipoprotein E-deficient (ApoE^−/−) mice, ApoE^−/−/5-LOX^−/−mice are protected against chronic HFD-induced liver injury, exhibiting significantly milder hepatic inflammation, insulin resistance, and macrophage infiltration.⁷⁶ 5-LOX expression is also upregulated in activated HSCs. Pharmacological inhibition of 5-LOX suppresses HSC activation and ameliorates carbon tetrachloride (CCl₄)- and MCD diet-induced liver fibrosis and liver injury.⁷⁹ 5-LOX can be inhibited by retinoic acid-related orphan receptor α (RORα). Melatonin directly inhibits HSC activation via RORα-mediated suppression of Alox5 expression.¹¹³

FIGURE 3.

The dual effects of lipoxygenases in metabolic dysfunction–associated steatotic liver disease and liver fibrosis. LTs and HETEs accelerate hepatocyte apoptosis, promote hepatic lipogenesis and inflammation, activate HSCs, and increase portal pressure. Moreover, LOXs directly induce ferroptosis, promoting hepatic lipogenesis. In contrast, SPMs drive M2 macrophage polarization, thereby attenuating hepatic inflammation and HSC activation. Created in BioRender. Ma, Y. (2025) https://BioRender.com/89ox4gn. Abbreviations: ACC1, acetyl-CoA carboxylase 1; CCL2, C-C motif chemokine ligand 2; cysLT, cysteinyl leukotriene; HETE, hydroxyeicosatetraenoic acid; LOX, lipoxygenase; LT, leukotriene; LX, lipoxin; MaR, maresin; RORα, retinoic acid-related orphan receptor α; ROS, reactive oxygen species; TGF-β1, transforming growth factor β1; TNF-α, tumor necrosis factor α.

LTB4 is a key product of the 5-LOX signaling pathway. LTB4 levels are significantly elevated in obese individuals and are crucial in pro-inflammatory cytokine production and insulin resistance.⁸⁰,¹¹⁴ Mechanistically, LTB4 directly enhances macrophage chemotaxis, stimulates inflammatory pathways, and impairs insulin-mediated suppression of hepatic glucose output in hepatocytes via an Gαi–c-Jun N-terminal Kinase 1 (JNK1)-dependent signaling pathway.¹¹⁴ In hepatocytes, the LTB4–BLT1 axis promotes hepatocyte lipogenesis by activating the protein kinase A (PKA)–inositol-requiring enzyme 1α (IRE1α)–spliced X-box-binding protein 1 (XBP1) signaling pathway, thereby upregulating lipogenic gene expression.⁸⁰,¹¹⁴ LTB4 and LTC4 activate HSCs via the ERK signaling pathway.⁷⁹ Moreover, CysLTs can decrease hepatic blood flow and increase portal pressure through their vasoconstrictor effects, potentially worsening liver fibrosis (Figure 3).³⁸ Collectively, these findings demonstrate the involvement of 5-LOX and its metabolites in MASLD and liver fibrosis. Targeting 5-LOX or specific LT receptors may be a promising therapeutic approach for treating these conditions.

12-LOX (platelet-type 12-LOX)

Emerging evidence highlights 12-LOX as a pivotal regulator of MASLD and liver fibrosis progression.¹⁶,⁴²,¹¹⁵ Hepatocyte-specific upregulation of ALOX12 markedly exacerbates hepatic lipid accumulation and inflammatory cascades.¹⁶ Mechanistically, 12-LOX directly targets acetyl-CoA carboxylase 1 (ACC1) to block Rab5-mediated ACC1 lysosomal degradation.⁴² The 12-LOX–ACC1 axis drives hepatic de novo lipogenesis and pro-inflammatory signaling, thereby accelerating MASH pathogenesis (Figure 3).⁴² Disruption of the 12-LOX–ACC1 axis via the small-molecule inhibitor IMA-1 demonstrates therapeutic efficacy.¹¹⁶ Moreover, HSCs exhibit elevated 12-LOX expression in MCD diet-induced mouse models.¹¹⁵ In addition to the 12-LOX–ACC1 axis, the 12-LOX–12-HETE axis is activated during hepatic ischemia-reperfusion injury and further enhanced in MASLD, correlating with increased HCC recurrence in steatotic livers. Specifically, 12-HETE promotes epithelial–mesenchymal transition and inflammatory activation via GPR31 (Figure 3).⁵¹,⁷⁷ Clinical studies have demonstrated a positive correlation between circulating 12-HETE levels and liver fibrosis severity.¹¹⁷ These findings identify the 12-LOX–ACC1/12-HETE signaling pathway as a promising therapeutic target for MASLD and liver fibrosis.

RORα is a key regulator of liver macrophage polarization. MaR1 is an endogenous ligand of RORα. In HFD-induced MASH mice, MaR1 enhances both the expression and transcriptional activity of RORα, thereby increasing the M2 polarization of liver macrophages and preventing inflammation. Furthermore, RORα upregulates the level of MaR1 by transcriptionally inducing the expression of 12-LOX (Figure 3).¹¹⁸ This MaR1/RORα/12-LOX autoregulatory circuit could offer potential therapeutic strategies for curing MASH.

12-/15-LOX (leukocyte-type 12-LOX)

Hepatic ALOX15/Alox15 gene expression and 12-/15-LOX metabolite levels significantly increased in mouse and human MASLD.⁷⁸,¹¹⁹,¹²⁰,¹²¹ 15-LOX upregulation enhances fatty acid uptake via the PPARγ–CD36 signaling pathway, leading to lipotoxicity and MASLD development.⁷⁸ In HFD-fed mice, 12-/15-LOX deficiency results in cullin2-mediated ubiquitination and degradation of xanthine oxidase, thereby reducing H₂O₂ production, de novo lipogenesis, and triglyceride biosynthesis.¹²⁰ Moreover, Alox15 is markedly upregulated in the livers of ApoE^−/− mice, which spontaneously develop MASLD secondary to hyperlipidemia. Genetic disruption of Alox15 protects ApoE^−/− mice from hepatic steatosis, insulin resistance, inflammation, and cell injury.¹¹⁹

While most studies indicate that ablation or inhibition of 12-/15-LOX protects against MASLD, some research has shown that reducing 12-/15-LOX-derived SPMs exacerbates severity.³⁸,¹²² LXA4, a bioactive metabolite of 12-/15-LOX, likely protects against liver diseases by reducing adipose tissue inflammation, modulating the M1/M2 macrophage phenotype, and promoting HFD-induced adipose autophagy.¹²² In addition, LXA4 reduces inflammatory cytokines and increases anti-inflammatory IL-10 levels in the systemic circulation.⁸¹ BML-11, an LX receptor agonist, ameliorates CCl₄-induced liver inflammation and fibrosis by inhibiting HSC activation and reducing TGF-β1 and PDGF levels (Figure 3).¹²³ These findings underscore the complexity of 12-/15-LOX and its metabolites in MASLD and liver fibrosis, emphasizing the need to understand their dual functions in both driving inflammation/fibrosis and promoting resolution.

LOX-related ferroptosis

Ferroptosis, a non-apoptotic form of regulated cell death, is characterized by iron-dependent necrosis primarily driven by extensive lipid peroxidation-mediated membrane damage.¹²⁴ LOXs have emerged as key mediators of ferroptosis.²⁶ Specifically, mammalian ALOX family members, such as 5-LOX, 15-LOX1, and 15-LOX2, are consistently shown to catalyze PUFA peroxidation, thereby promoting ferroptosis (Figure 3).²⁶,⁴¹,¹²⁴ Notably, while LOX catalysis inhibition fails to rescue cells from ferroptosis induced by common ferroptosis inducers, radical-trapping antioxidants can effectively suppress ferroptosis. This finding suggests that although LOXs are not essential for ferroptosis execution, they may contribute to its initiation by increasing cellular lipid hydroperoxide levels, which subsequently propagate lipid autoxidation.²⁶

Ferroptosis is intricately linked to various cellular processes, including iron homeostasis, redox balance, lipid metabolism, and glutaminolysis.¹²⁵ Elevated AA metabolism contributes to ferroptosis, and lipid ROS produced by ferroptosis trigger lipid accumulation, thereby promoting the progression of MASLD.¹²⁶ Ferroptosis inhibitors ameliorate MCD diet-induced liver injury by suppressing hepatic lipotoxicity (Figure 3).¹²⁶ The primary ferroptosis-promoting pathway in MASLD involves the acyl-CoA synthetase long-chain family member 4 (ACSL4)–15-LOX axis.¹²⁷ The small scaffolding protein PEBP1 regulates ferroptotic cell death by binding to 15-LOXs and facilitating the generation of lipid peroxides. Inhibiting the formation of the PEBP1/15-LOX complex could lead to novel anti-ferroptosis strategies.¹²⁸ For example, ferroLOXINs are anti-ferroptotic molecules that selectively target the 15-LOX2/PEBP1 complex rather than 15-LOX2 alone, without affecting the biosynthesis of pro-inflammatory or anti-inflammatory lipid mediators.¹²⁹

Role of CYP enzymes and their products in MASLD and liver fibrosis

CYP enzymes, notably those with epoxygenase and ω-hydroxylase activity, metabolize AA and contribute to the pathogenesis of MASLD and liver fibrosis. Epoxygenase converts AA to EETs, which exert hepatoprotective effects by alleviating inflammatory responses and oxidative stress, thereby attenuating hepatic steatosis and fibrosis.¹⁶,⁸² Conversely, ω-hydroxylase is implicated in promoting MASLD and liver fibrosis.¹⁹

Epoxygenase

CYP2J2, CYP2C8, CYP2C9, CYP1A2, and CYP3A4 function as epoxygenases in the human liver, generating EETs with anti-inflammatory, antisteatotic, and antifibrotic properties.¹⁶ Both mice with MCD diet-induced MASH and patients with biopsy-confirmed MASH exhibit significantly higher free EET concentrations compared with healthy controls.⁵⁶ Circulating EETs are inversely correlated with MASLD prevalence and histological severity, emerging as potential MASH biomarkers.¹³⁰ Specifically, 11,12-EET and 14,15-EET, products of AA catalyzed by CYP2J2, mitigate lipotoxicity-related inflammation and oxidative stress in hepatocytes and macrophages.⁸²,⁸³ EET intervention reduces fatty acid accumulation, MASLD, and fibrosis by upregulating HO-1–PGC1α.⁸⁴ Furthermore, endothelial-specific overexpression of CYP2J2 alleviates hepatic steatosis, inflammation, and oxidative stress by inhibiting NF-κB/JNK signaling and enhancing antioxidant defenses.⁸² Consistent with these findings, Ephx2 (encoding sEH) knockout in mice increases EET levels and attenuates hepatic steatosis, inflammation, injury, and systemic glucose intolerance (Figure 4).⁵⁶

FIGURE 4.

The main biological functions of epoxygenase and ω-hydroxylase in metabolic dysfunction–associated steatotic liver disease and liver fibrosis. Epoxygenases convert AA to EETs, which mitigate hepatic steatosis and fibrosis progression. However, sEH can hydrolyze EETs, thereby diminishing their protective effects in the liver. ω-Hydroxylase family members contribute to the progression of MASLD and liver fibrosis. Created in BioRender. Ma, Y. (2025) https://BioRender.com/fjz4iim. Abbreviations: AA, arachidonic acid; COX-2, cyclooxygenase-2; CYP, cytochrome P450; EET, epoxyeicosatrienoic acid; ER, endoplasmic reticulum; FAT, fatty acid translocase; ROS, reactive oxygen species; sEH, soluble epoxide hydrolase; TGF-β, transforming growth factor β; TNF-α, tumor necrosis factor α.

Dysregulation of the AA–CYP–EET–sEH signaling pathway is a key pathological feature of MASLD and liver fibrosis.¹³¹,¹³²,¹³³,¹³⁴ sEH inhibition attenuates MASLD and liver fibrosis by reducing inflammation, oxidative stress, and ER stress.¹³³ PTUPB ameliorates liver fibrosis and portal hypertension by downregulating the sEH–COX-2–TGF-β signaling pathway (Figure 4).¹³² In addition, the sEH inhibitor t-TUCB decreases portal hypertension and attenuates hepatic endothelial dysfunction and liver fibrosis in cirrhotic rats.¹³³,¹³⁴ Thus, targeting the AA–CYP–EET–sEH signaling pathway holds promise for MASLD and therapy. However, the roles of other AA–CYP signaling pathways in MASLD and liver fibrosis remain to be elucidated.

ω-Hydroxylase

The expression of the ω-hydroxylase gene family CYP4A is upregulated in the livers of patients with MASLD and diet-induced MASLD mouse models, with a significant increase in CYP4A11 expression.¹⁹,⁵⁸ The plasma CYP4A11 concentration is strongly correlated with lipid peroxidation biomarkers.¹³⁵ In HepG2 cells stimulated with free fatty acids, CYP4A11 induces the production of ROS and the release of pro-inflammatory cytokines, including TNF-α, IL-1β, and IL-6, suggesting that CYP4A11 may metabolize free fatty acids to promote ROS generation and inflammatory responses, thereby accelerating MASLD progression.¹³⁵ In addition, CYP4A14 significantly exacerbates HFD- and MCD-induced liver lipid accumulation and MASH progression by increasing fatty acid translocase (FAT/CD36) expression (Figure 4).¹⁹ Furthermore, 20-HETE, a prominent product of CYP4A, contributes significantly to obesity-related insulin resistance and hyperlipidemia by disrupting insulin signaling in differentiated adipocytes.⁸⁵ Collectively, CYP4A family members significantly contribute to MASLD progression via ROS generation, inflammation, and lipid metabolism. Future research should clarify CYP4A-regulated signaling pathways and explore CYP4A inhibitors as potential MASLD therapeutics.

Role of gut microbiota–related AA metabolism in MASLD and liver fibrosis

Intestinal barrier damage in MASLD leads to alterations in the gut microbiota via gut microbiota-dependent AA metabolism, thereby accelerating hepatic inflammation and the progression of MASLD.¹³⁶,¹³⁷,¹³⁸ Specifically, fructose corn syrup increases the abundance of Muribaculaceae and enhances microbial functions involved in AA metabolism, resulting in elevated PGs and significantly increased fat deposition, hepatic steatosis, and liver and intestinal inflammatory damage.¹³⁸ Comparative analysis reveals that obesity-resistant mice exhibit lower body weights, liver weights, adipose accumulation, and pro-inflammatory cytokine levels compared with obesity-prone mice. The genera Allobaculum, SMB53, Desulfovibrio, and Clostridium are enriched in obesity-prone mice, whereas Streptococcus, Odoribacter, and Leuconostoc are enriched in obesity-resistant mice, with significantly altered bacteria and AA metabolites showing strong correlations.¹³⁹ In addition, pyruvate derived from Lachnospiraceae suppresses the expression of hepatic forkhead box O3 (FOXO3) and ALOX15, thereby reducing liver inflammation, susceptibility to ferroptosis, and the inflammatory immune cell infiltration.¹⁴⁰

Obeticholic acid (OCA) has been examined for the treatment of MASLD, but it has shown unsatisfactory antifibrotic effects and a low response rate in phase III clinical trials.¹⁴¹ Mechanistically, OCA-modulated changes in the gut microbiota induce lipid peroxidation and impair its antifibrotic effect. Bacteroides enriched by OCA induce liver ROS accumulation. OCA reduces triglycerides containing PUFA levels, whereas elevates free PUFAs and phosphatidylethanolamines containing PUFA, which are susceptible to oxidation to lipid peroxides (notably AA-derived 12-hydroxyheptadecatrienoic acid), inducing hepatocyte ferroptosis and activating HSCs.¹⁴² These findings demonstrate that the gut microbiota–related AA metabolic pathway plays a crucial role in the progression of MASLD and liver fibrosis. Restoring the disrupted microbiota balance can normalize aberrant AA metabolism, which holds promise as a potential therapeutic strategy for MASLD and liver fibrosis.

AA METABOLIC ENZYMES AND METABOLITES AS POTENTIAL THERAPEUTIC TARGETS IN MASLD AND LIVER FIBROSIS

The AA signaling pathway regulates the occurrence and resolution of inflammation and is closely related to the pathophysiology of MASLD and liver fibrosis. Given the pivotal roles of the AA metabolic pathway in MASLD and liver fibrosis, the development of drugs or biological products that target AA metabolic enzymes and metabolites holds significant potential. In recent years, AA derivatives, including COX-1/2 inhibitors, LOX inhibitors, and CYP inhibitors or agonists, have been extensively utilized in preclinical studies of MASLD and liver fibrosis (Table 3).

TABLE 3.

Preclinical drug research targeting the arachidonic acid metabolic pathway

Drugs	Targets	Effects	Reference
Celecoxib	COX-2 inhibitor	Reduces fibrosis in MASH and alleviates liver fibrosis and portal hypertension	72,75
Nimesulide	COX-2 inhibitor	Suppresses MASLD and hepatic insulin resistance	70
Zileuton	5-LOX inhibitor	Reverses elevated transaminase activities and MASLD activity score, attenuates liver fibrosis	79,112
Zileuton liposomes	5-LOX inhibitor	Improves CCl4- and MCD diet-induced hepatic fibrosis and liver injury	79
ML355	12-LOX inhibitor	Reduces HCC recurrence in fatty livers	77
IMA-1	12-LOX inhibitor	Alleviates MASH	116
Gigantol	12-LOX inhibitor	Alleviates liver inflammation	143
CP105696	BLT1 antagonist	Inhibits hepatic steatosis in HFD-induced and genetic obesity	114
BML-11	LX receptor agonist	Alleviates MASLD and fibrosis	123
C418	CYP4A inhibitor	Ameliorates MASLD	144
t-TUCB	sEH inhibitor	Alleviates liver fibrosis and portal hypertension	133,134
PTUPB	COX-2/sEH dual inhibitor	Alleviates MASLD, liver fibrosis and portal hypertension	24,131,132
Dihydroartemisinin	COX-2/5-LOX inhibitor	Mitigates liver inflammation	145
Baicalein	COX-2, CYP2E1 inhibitor	Attenuates hepatic steatosis, inflammation and fibrosis	146
Berberine	PLA2, COX-2, 5-LOX inhibitor	Prevents MASLD and liver fibrosis progression	147

Abbreviations: BLT1, leukotriene B4 receptor 1; CCl₄, carbon tetrachloride; COX, cyclooxygenase; CYP, cytochrome P450; HFD, high-fat diet; LOX, lipoxygenase; LX, lipoxin; MCD, methionine- and choline-deficient; PLA2, phospholipase A2; sEH, soluble epoxide hydrolase.

COX-2 inhibitors

Celecoxib, a selective COX-2 inhibitor, has demonstrated multiple beneficial effects on MASLD and liver fibrosis in experimental animal models. Specifically, celecoxib attenuates hepatocyte apoptosis,¹⁰² mitigates inflammation and oxidative stress,¹⁰⁷,¹⁴⁸ inhibits epithelial–mesenchymal transition⁹³ and angiogenesis,⁷⁵ restores gut barrier integrity,⁷³ and counteracts fibrogenesis,⁹⁴ thereby protecting against MASLD and liver fibrosis. Similarly, nimesulide, another selective COX-2 inhibitor, suppresses MASLD and hepatic insulin resistance by regulating PPARγ.⁷⁰ These findings suggest that COX-2 inhibitors, particularly celecoxib, hold promise as therapeutic agents for MASLD and liver fibrosis. However, COX-2 inhibition may lead to side effects, and further extensive clinical research is needed to verify the efficacy and safety in clinical applications.

LOX inhibitors

LOXs and their catalyzed metabolites are implicated in both the progression and resolution of MASLD and liver fibrosis. Pharmacological inhibition of LOXs and their related metabolites has been shown to alleviate liver steatosis, inflammation, and fibrosis.⁷⁹,⁸⁰,¹¹⁴ However, LOX-derived SPMs play a crucial role in MASLD and liver fibrosis resolution.³⁴,¹⁴⁹

Zileuton, a clinically approved 5-LOX inhibitor, delays HFD-induced MASLD progression¹¹² and CCl₄- and MCD diet-induced liver fibrosis progression.⁷⁹ Nevertheless, zileuton can cause hepatocyte damage and is associated with hepatotoxicity, as evidenced by elevated transaminase levels.¹⁵⁰ To circumvent this issue, targeted delivery of zileuton to HSCs via cRGDyK-modified liposomes, which bind to integrin αvβ3 expressed by HSCs, strongly alleviates MCD diet- and CCl₄-induced liver fibrosis without off-target effects.⁷⁹

12-LOX and 15-LOX are emerging as potential therapeutic targets for MASLD and liver fibrosis. Gigantol alleviates liver inflammation partly by inhibiting the JNK–cPLA2–12-LOX signaling pathway.¹⁴³ The inhibition of ALOX12–12-HETE by ML355 reduces HCC recurrence in fatty livers.⁷⁷ IMA-1, a novel small-molecule inhibitor of 12-LOX, is more stable and has a longer half-life than its parent compound ML355. IMA-1 treatment disrupts the ALOX12–ACC1 interaction and facilitates ACC1 protein degradation. Specifically, inhibiting the ALOX12–ACC1 interaction not only effectively inhibits MASLD progression but also prevents the induction of hyperlipidemia as a side effect, indicating that IMA-1 is a promising therapeutic agent for MASLD.¹¹⁶ Since ferroptosis is involved in the pathophysiology of MASLD and liver fibrosis, inhibiting ferroptosis by LOX inhibitors or anti-ferroptosis FerroLOXINs could be an effective therapeutic approach.¹²⁹

CYP4A and sEH inhibitors

Given the detrimental effects of 20-HETE and the protective effects of EETs, targeting HETE and EET regulators holds therapeutic promise for MASLD and liver fibrosis. C418, a CYP4A inhibitor, ameliorates steatosis and MASLD by inhibiting ER stress/oxidative stress, lipotoxicity, inflammation, and fibrosis.¹⁴⁴ In addition, t-TUCB, an sEH inhibitor that decreases the metabolism of EETs, improves portal hypertension and liver fibrosis.¹³³,¹³⁴ However, research in this area remains limited, and further investigation is warranted.

Natural products

Natural products also exhibit protective effects against MASLD and liver fibrosis by inhibiting multiple AA metabolic enzymes. For example, dihydroartemisinin modulates AA metabolism and alleviates liver inflammation by inhibiting the activation of COX-2 and 5-LOX.¹⁴⁵ Similarly, baicalin attenuates hepatic steatosis, inflammation, and fibrosis by suppressing pro-inflammatory COX-2 and CYP2E1.¹⁴⁶ Berberine reduces hepatic lipid accumulation and prevents the progression of MASLD and liver fibrosis by affecting the activity of metabolic enzymes such as PLA2, 5-LOX, and COX-2.¹⁴⁷ As natural products show less toxicity and promising efficacy in treating MASLD and liver fibrosis, further studies are needed to explore more AA metabolic-targeting natural products.

CLINICAL STUDIES TARGETING THE AA METABOLIC PATHWAY IN MASLD

Nonsteroidal anti-inflammatory drugs (NSAIDs), such as aspirin, celecoxib, and rofecoxib, are well-known COX-1 and/or COX-2 inhibitors used to alleviate pain, fever, and inflammation. In a clinical trial, aspirin has been demonstrated to reduce intrahepatic lipid content (NCT04031729). Celecoxib, approved for treating arthritis,¹⁵¹ has been shown to enhance insulin sensitivity in both obese and healthy individuals.¹⁵²,¹⁵³ Meloxicam, another COX-2 inhibitor, is currently being evaluated in a phase IV clinical trial (NCT02298218) for its potential effects on liver cirrhosis associated with biliary atresia, with results yet to be disclosed. A multicentre, double-blind, randomized, placebo-controlled trial demonstrated that sequential treatment with oral parecoxib followed by parenteral imrecoxib effectively suppressed systemic inflammation.¹⁵⁴ Misoprostol, a PGE1 analog, is being investigated in a phase II clinical trial (NCT05804305) for its potential to alleviate MASLD.¹⁵⁵ Moreover, selective COX-2 inhibitors can improve clinical outcomes and liver function in cirrhotic patients with good safety in a retrospective study.¹⁵⁶ These translational medicine studies shed light on COX-2 inhibitors as a new clinical treatment for MASLD and liver cirrhosis.

In a clinical trial (NCT05554224), plasma concentrations of fatty acids and oxylipins were significantly higher in patients with severe obesity compared with their metabolically healthier overweight/obese counterparts. Before surgery, oxylipins derived from LOX activity, such as 12-HETE, 11-hydroxydocosahexaenoic acid (11-HDoHE), 14-HDoHE, and 12-hydroxyeicosapentaenoic acid (12-HEPE), were predominant. However, one year following laparoscopic sleeve gastrectomy, oxylipins are mainly from the COX activity, particularly pro-inflammatory prostanoids like TXB2, PGE2, PGD2, and 12-hydroxyheptadecatrienoate (12-HHTrE). This transformation appears to be linked to a reduction in adiposity.¹⁵⁷

To date, no clinical trials have directly targeted CYP enzymes or their direct products in MASLD and liver fibrosis. However, liver cirrhosis differentially affects the activity of CYP isoforms, necessitating consideration for adjusting the dose of drugs used in cirrhotic patients (NCT03337945).¹⁵⁸ In a phase II clinical trial (NCT00847899), the safety and efficacy of AR9281, a novel sEH enzyme inhibitor, were evaluated for improving glucose metabolism, insulin resistance, and blood pressure in patients with impaired glucose tolerance and mild-to-moderate hypertension, but the outcomes remain undeclared.

PERSPECTIVES AND CONCLUSIONS

Despite accumulating evidence implicating AA metabolism in MASLD and liver fibrosis, several gaps and challenges remain. First, the specific liver cell types in which COXs, LOXs, and CYP enzymes primarily affect the progression or resolution of MASLD and liver fibrosis are not fully understood. Studies have shown that COX-2 is predominantly active in HSCs, KCs, and LSECs, where it promotes hepatic inflammation, lipid accumulation, fibrogenesis, and angiogenesis. Under normal physiological conditions, mature hepatocytes lack detectable COX-2 activity. However, recent studies have demonstrated that hepatocyte-specific overexpression of COX-2 attenuates MASLD and liver fibrosis. The cellular localization of LOXs and CYP enzymes in the liver remains unclear. Therefore, advanced techniques, such as real-time and cell-specific lipidomic analysis, spatial metabolomics, and matrix-assisted laser desorption ionization mass spectrometry imaging (MALDI-MSI), are needed to map the cellular and tissue distribution of AA metabolic enzymes and their metabolites, thereby deepening our understanding of AA metabolic complexity and its crosstalk in the liver. Second, substantial evidence links the gut microbiota to MASLD and liver fibrosis, warranting further investigation into the relationship between AA and gut microbiota dynamics. Third, the heterogeneity of MASLD presents a complex clinical picture, complicating the establishment of a unified pathogenic mechanism. Further research is needed to delineate the specific roles of AA metabolites and signaling pathways in various stages of MASLD progression, ranging from simple steatosis to advanced fibrosis and cirrhosis. Fourth, translating basic research findings into effective clinical therapies continues to pose a substantial challenge. Well-designed clinical trials and pharmacological efficacy studies could offer direct evidence supporting the potential of AA metabolic pathways in preventing and treating MASLD and liver fibrosis. Future research should focus on addressing the gaps in our understanding of AA metabolism, elucidating its interactions with the gut microbiota, and translating these findings into effective clinical strategies.

In conclusion, AA metabolism, mediated by diverse enzymatic pathways, plays a crucial role in the pathogenesis of MASLD and liver fibrosis, influencing multiple aspects of liver pathology. AA and its metabolites affect hepatic inflammation, lipid accumulation, fibrogenesis, and various cellular processes, including apoptosis, pyroptosis, autophagy, cellular senescence, and ferroptosis. Given the dual roles of AA and its metabolites in either promoting or alleviating MASLD and liver fibrosis, many inhibitors and agonists have been investigated as potential therapeutic targets.